[addresses/hyphens] adding some methods to hyphenate/dehyphenate place names at random

2016-10-18 19:10:31 -04:00
parent 7e007a49ab
commit 72e7d3ff5b
2 changed files with 40 additions and 0 deletions
--- a/resources/parser/default.yaml
+++ b/resources/parser/default.yaml
@@ -108,6 +108,10 @@ category:
    drop_places_probability: 0.1 # drop place names
    drop_postcode_probability: 0.3 # drop postal code

+places:
+    hyphenate_multiword_probability: 0.005
+    remove_hyphen_probability: 0.5
+
 boundaries:
    abbreviate_toponym_probability: 0.35